| Message ID | 20191118230254.2615942-6-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [01/19] drm/i915/selftests: Force bonded submission to overlap | expand |
On 18/11/2019 23:02, Chris Wilson wrote: > In order to avoid some nasty mutex inversions, commit 09c5ab384f6f > ("drm/i915: Keep rings pinned while the context is active") allowed the > intel_ring unpinning to be run concurrently with the next context > pinning it. Thus each step in intel_ring_unpin() needed to be atomic and > ordered in a nice onion with intel_ring_pin() so that the lifetimes > overlapped and were always safe. > > Sadly, a few steps in intel_ring_unpin() were overlooked, such as > closing the read/write pointers of the ring and discarding the > intel_ring.vaddr, as these steps were not serialised with > intel_ring_pin() and so could leave the ring in disarray. > > Fixes: 09c5ab384f6f ("drm/i915: Keep rings pinned while the context is active") > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> > --- > drivers/gpu/drm/i915/gt/intel_ring.c | 13 ++++--------- > 1 file changed, 4 insertions(+), 9 deletions(-) > > diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c > index ece20504d240..374b28f13ca0 100644 > --- a/drivers/gpu/drm/i915/gt/intel_ring.c > +++ b/drivers/gpu/drm/i915/gt/intel_ring.c > @@ -57,9 +57,10 @@ int intel_ring_pin(struct intel_ring *ring) > > i915_vma_make_unshrinkable(vma); > > - GEM_BUG_ON(ring->vaddr); > - ring->vaddr = addr; > + /* Discard any unused bytes beyond that submitted to hw. */ > + intel_ring_reset(ring, ring->emit); > > + ring->vaddr = addr; > return 0; > > err_ring: > @@ -85,20 +86,14 @@ void intel_ring_unpin(struct intel_ring *ring) > if (!atomic_dec_and_test(&ring->pin_count)) > return; > > - /* Discard any unused bytes beyond that submitted to hw. */ > - intel_ring_reset(ring, ring->emit); > - > i915_vma_unset_ggtt_write(vma); > if (i915_vma_is_map_and_fenceable(vma)) > i915_vma_unpin_iomap(vma); > else > i915_gem_object_unpin_map(vma->obj); > > - GEM_BUG_ON(!ring->vaddr); > - ring->vaddr = NULL; > - > - i915_vma_unpin(vma); > i915_vma_make_purgeable(vma); > + i915_vma_unpin(vma); > } > > static struct i915_vma *create_ring_vma(struct i915_ggtt *ggtt, int size) > Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Regards, Tvrtko
diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c index ece20504d240..374b28f13ca0 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring.c +++ b/drivers/gpu/drm/i915/gt/intel_ring.c @@ -57,9 +57,10 @@ int intel_ring_pin(struct intel_ring *ring) i915_vma_make_unshrinkable(vma); - GEM_BUG_ON(ring->vaddr); - ring->vaddr = addr; + /* Discard any unused bytes beyond that submitted to hw. */ + intel_ring_reset(ring, ring->emit); + ring->vaddr = addr; return 0; err_ring: @@ -85,20 +86,14 @@ void intel_ring_unpin(struct intel_ring *ring) if (!atomic_dec_and_test(&ring->pin_count)) return; - /* Discard any unused bytes beyond that submitted to hw. */ - intel_ring_reset(ring, ring->emit); - i915_vma_unset_ggtt_write(vma); if (i915_vma_is_map_and_fenceable(vma)) i915_vma_unpin_iomap(vma); else i915_gem_object_unpin_map(vma->obj); - GEM_BUG_ON(!ring->vaddr); - ring->vaddr = NULL; - - i915_vma_unpin(vma); i915_vma_make_purgeable(vma); + i915_vma_unpin(vma); } static struct i915_vma *create_ring_vma(struct i915_ggtt *ggtt, int size)
In order to avoid some nasty mutex inversions, commit 09c5ab384f6f ("drm/i915: Keep rings pinned while the context is active") allowed the intel_ring unpinning to be run concurrently with the next context pinning it. Thus each step in intel_ring_unpin() needed to be atomic and ordered in a nice onion with intel_ring_pin() so that the lifetimes overlapped and were always safe. Sadly, a few steps in intel_ring_unpin() were overlooked, such as closing the read/write pointers of the ring and discarding the intel_ring.vaddr, as these steps were not serialised with intel_ring_pin() and so could leave the ring in disarray. Fixes: 09c5ab384f6f ("drm/i915: Keep rings pinned while the context is active") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> --- drivers/gpu/drm/i915/gt/intel_ring.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-)